The objective of this epidemiologic study is to answer the question whether gastric resection in benign ulcer disease is a risk factor for gastric cancer: and if so, to which extent this risk is modified by duration of follow-up, age at surgery, sex, localization of ulcer disease and type of enteral anastomosis (BI bs. BII). The answer to these questions are fundamental for the understanding of possible carcinogenic effects of biochemical changes in the gastric remnant, for knowledge of the natural history of allegedly premalignant changes in the gastric mucosa and for a cost-effective design of programs for endoscopic surveillance after gastric resection. The aim is to overcome some of the major methodological difficulties underlying the contradictory findings in this field through the following design: 1. Recruitment of cohort comprising virtually all patients subjected to gastric resection in a defined geographic area during the period 1950-58. The size of the cohort - about 6,000 patients - will enable establishment of a relative risk of 1.5 or larger (alpha 0.05, beta 0.15). 2. The cohort will be based on data gathered from hospital records at the surgical clinics, supplemented with information from parish authorities and population register including national registration numbers. 3. Computerized linkage of these numbers of the cohort members to the national register covering emigration and causes of death will enable calculation of person-years at risk by age, sex, calendar year, duration of observation etc. Based on these data and the annual age- and sex specific incidence of gastric cancer in the area of study, the expected cancer incidence during the period 1958-82 will be calculated. 4. The observed cancer incidence in the cohort will be obtained by a linkage to the entire Swedish Cancer Registry. A large and population based cohort, long-term observation (0-32 completed years), close to complete follow-up and the availability of reliable incidence rates for gastric cancer in the studied population will enable advancement of knowledge and thus offer a more solid basis for etiological hypotheses and intervention strategies in the future.